[Cosmic][PATCH 0/1] dax: remove VM_MIXEDMAP for fsdax and device dax

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[Cosmic][PATCH 0/1] dax: remove VM_MIXEDMAP for fsdax and device dax

Joseph Salisbury-3
BugLink: https://bugs.launchpad.net/bugs/1789146

== Cosmic Justification ==
Intel is requesting this commit in Cosmic.  It resolves a bug that causes an
 assertion failure with util_range_rw using libpmemlog.

Commit e1fb4a086495 is in mailine as of v4.19-rc1.

== Fix ==
e1fb4a086495 ("dax: remove VM_MIXEDMAP for fsdax and device dax")

Dave Jiang (1):
  dax: remove VM_MIXEDMAP for fsdax and device dax

 drivers/dax/device.c | 2 +-
 fs/ext2/file.c       | 1 -
 fs/ext4/file.c       | 2 +-
 fs/xfs/xfs_file.c    | 2 +-
 mm/hmm.c             | 6 ++++--
 mm/huge_memory.c     | 4 ++--
 mm/ksm.c             | 3 +++
 mm/memory.c          | 6 ++++++
 mm/migrate.c         | 3 ++-
 mm/mlock.c           | 3 ++-
 mm/mmap.c            | 9 +++++----
 11 files changed, 27 insertions(+), 14 deletions(-)

--
2.7.4


--
kernel-team mailing list
[hidden email]
https://lists.ubuntu.com/mailman/listinfo/kernel-team
Reply | Threaded
Open this post in threaded view
|

[Cosmic][PATCH 1/1] dax: remove VM_MIXEDMAP for fsdax and device dax

Joseph Salisbury-3
From: Dave Jiang <[hidden email]>

BugLink: https://bugs.launchpad.net/bugs/1789146

This patch is reworked from an earlier patch that Dan has posted:
https://patchwork.kernel.org/patch/10131727/

VM_MIXEDMAP is used by dax to direct mm paths like vm_normal_page() that
the memory page it is dealing with is not typical memory from the linear
map.  The get_user_pages_fast() path, since it does not resolve the vma,
is already using {pte,pmd}_devmap() as a stand-in for VM_MIXEDMAP, so we
use that as a VM_MIXEDMAP replacement in some locations.  In the cases
where there is no pte to consult we fallback to using vma_is_dax() to
detect the VM_MIXEDMAP special case.

Now that we have explicit driver pfn_t-flag opt-in/opt-out for
get_user_pages() support for DAX we can stop setting VM_MIXEDMAP.  This
also means we no longer need to worry about safely manipulating vm_flags
in a future where we support dynamically changing the dax mode of a
file.

DAX should also now be supported with madvise_behavior(), vma_merge(),
and copy_page_range().

This patch has been tested against ndctl unit test.  It has also been
tested against xfstests commit: 625515d using fake pmem created by
memmap and no additional issues have been observed.

Link: http://lkml.kernel.org/r/152847720311.55924.16999195879201817653.stgit@...
Signed-off-by: Dave Jiang <[hidden email]>
Acked-by: Dan Williams <[hidden email]>
Cc: Jan Kara <[hidden email]>
Signed-off-by: Andrew Morton <[hidden email]>
Signed-off-by: Linus Torvalds <[hidden email]>
(cherry picked from commit e1fb4a0864958fac2fb1b23f9f4562a9f90e3e8f)
Signed-off-by: Joseph Salisbury <[hidden email]>
---
 drivers/dax/device.c | 2 +-
 fs/ext2/file.c       | 1 -
 fs/ext4/file.c       | 2 +-
 fs/xfs/xfs_file.c    | 2 +-
 mm/hmm.c             | 6 ++++--
 mm/huge_memory.c     | 4 ++--
 mm/ksm.c             | 3 +++
 mm/memory.c          | 6 ++++++
 mm/migrate.c         | 3 ++-
 mm/mlock.c           | 3 ++-
 mm/mmap.c            | 9 +++++----
 11 files changed, 27 insertions(+), 14 deletions(-)

diff --git a/drivers/dax/device.c b/drivers/dax/device.c
index 77fff4d..6fd4608 100644
--- a/drivers/dax/device.c
+++ b/drivers/dax/device.c
@@ -495,7 +495,7 @@ static int dax_mmap(struct file *filp, struct vm_area_struct *vma)
  return rc;
 
  vma->vm_ops = &dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
  return 0;
 }
 
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 047c327..28b2609 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -126,7 +126,6 @@ static int ext2_file_mmap(struct file *file, struct vm_area_struct *vma)
 
  file_accessed(file);
  vma->vm_ops = &ext2_dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP;
  return 0;
 }
 #else
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 7f80233..69d65d4 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -374,7 +374,7 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma)
  file_accessed(file);
  if (IS_DAX(file_inode(file))) {
  vma->vm_ops = &ext4_dax_vm_ops;
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
  } else {
  vma->vm_ops = &ext4_file_vm_ops;
  }
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index a3e7767..d14d0f7 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -1171,7 +1171,7 @@ xfs_file_mmap(
  file_accessed(filp);
  vma->vm_ops = &xfs_file_vm_ops;
  if (IS_DAX(file_inode(filp)))
- vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
+ vma->vm_flags |= VM_HUGEPAGE;
  return 0;
 }
 
diff --git a/mm/hmm.c b/mm/hmm.c
index f9d1d89..e1ea093 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -676,7 +676,8 @@ int hmm_vma_get_pfns(struct hmm_range *range)
  return -EINVAL;
 
  /* FIXME support hugetlb fs */
- if (is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL)) {
+ if (is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL) ||
+ vma_is_dax(vma)) {
  hmm_pfns_special(range);
  return -EINVAL;
  }
@@ -849,7 +850,8 @@ int hmm_vma_fault(struct hmm_range *range, bool block)
  return -EINVAL;
 
  /* FIXME support hugetlb fs */
- if (is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL)) {
+ if (is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL) ||
+ vma_is_dax(vma)) {
  hmm_pfns_special(range);
  return -EINVAL;
  }
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4e5041c..7a66941 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -762,11 +762,11 @@ vm_fault_t vmf_insert_pfn_pmd(struct vm_area_struct *vma, unsigned long addr,
  * but we need to be consistent with PTEs and architectures that
  * can't support a 'special' bit.
  */
- BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)));
+ BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) &&
+ !pfn_t_devmap(pfn));
  BUG_ON((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) ==
  (VM_PFNMAP|VM_MIXEDMAP));
  BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags));
- BUG_ON(!pfn_t_devmap(pfn));
 
  if (addr < vma->vm_start || addr >= vma->vm_end)
  return VM_FAULT_SIGBUS;
diff --git a/mm/ksm.c b/mm/ksm.c
index c7f9a1a..a13c45c 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -2434,6 +2434,9 @@ int ksm_madvise(struct vm_area_struct *vma, unsigned long start,
  VM_HUGETLB | VM_MIXEDMAP))
  return 0; /* just ignore the advice */
 
+ if (vma_is_dax(vma))
+ return 0;
+
 #ifdef VM_SAO
  if (*vm_flags & VM_SAO)
  return 0;
diff --git a/mm/memory.c b/mm/memory.c
index 86d4329..356649f 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -871,6 +871,10 @@ struct page *_vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
  return NULL;
  }
  }
+
+ if (pte_devmap(pte))
+ return NULL;
+
  print_bad_pte(vma, addr, pte, NULL);
  return NULL;
  }
@@ -935,6 +939,8 @@ struct page *vm_normal_page_pmd(struct vm_area_struct *vma, unsigned long addr,
  }
  }
 
+ if (pmd_devmap(pmd))
+ return NULL;
  if (is_zero_pfn(pfn))
  return NULL;
  if (unlikely(pfn > highest_memmap_pfn))
diff --git a/mm/migrate.c b/mm/migrate.c
index 8c0af0f..4a83268 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2951,7 +2951,8 @@ int migrate_vma(const struct migrate_vma_ops *ops,
  /* Sanity check the arguments */
  start &= PAGE_MASK;
  end &= PAGE_MASK;
- if (!vma || is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL))
+ if (!vma || is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_SPECIAL) ||
+ vma_is_dax(vma))
  return -EINVAL;
  if (start < vma->vm_start || start >= vma->vm_end)
  return -EINVAL;
diff --git a/mm/mlock.c b/mm/mlock.c
index 74e5a65..41cc47e 100644
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -527,7 +527,8 @@ static int mlock_fixup(struct vm_area_struct *vma, struct vm_area_struct **prev,
  vm_flags_t old_flags = vma->vm_flags;
 
  if (newflags == vma->vm_flags || (vma->vm_flags & VM_SPECIAL) ||
-    is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm))
+    is_vm_hugetlb_page(vma) || vma == get_gate_vma(current->mm) ||
+    vma_is_dax(vma))
  /* don't set VM_LOCKED or VM_LOCKONFAULT and don't count */
  goto out;
 
diff --git a/mm/mmap.c b/mm/mmap.c
index a31c3e1..d4f4e13 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1796,11 +1796,12 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
 
  vm_stat_account(mm, vm_flags, len >> PAGE_SHIFT);
  if (vm_flags & VM_LOCKED) {
- if (!((vm_flags & VM_SPECIAL) || is_vm_hugetlb_page(vma) ||
- vma == get_gate_vma(current->mm)))
- mm->locked_vm += (len >> PAGE_SHIFT);
- else
+ if ((vm_flags & VM_SPECIAL) || vma_is_dax(vma) ||
+ is_vm_hugetlb_page(vma) ||
+ vma == get_gate_vma(current->mm))
  vma->vm_flags &= VM_LOCKED_CLEAR_MASK;
+ else
+ mm->locked_vm += (len >> PAGE_SHIFT);
  }
 
  if (file)
--
2.7.4


--
kernel-team mailing list
[hidden email]
https://lists.ubuntu.com/mailman/listinfo/kernel-team
Reply | Threaded
Open this post in threaded view
|

APPLIED: [Cosmic][PATCH 0/1] dax: remove VM_MIXEDMAP for fsdax and device dax

Thadeu Lima de Souza Cascardo-3
In reply to this post by Joseph Salisbury-3
Applied to cosmic master-next branch.

Thanks.
Cascardo.

Applied-to: cosmic/master-next

--
kernel-team mailing list
[hidden email]
https://lists.ubuntu.com/mailman/listinfo/kernel-team